CN112330561A - Medical image segmentation method based on interactive foreground extraction and information entropy watershed - Google Patents
Medical image segmentation method based on interactive foreground extraction and information entropy watershed Download PDFInfo
- Publication number
- CN112330561A CN112330561A CN202011226634.0A CN202011226634A CN112330561A CN 112330561 A CN112330561 A CN 112330561A CN 202011226634 A CN202011226634 A CN 202011226634A CN 112330561 A CN112330561 A CN 112330561A
- Authority
- CN
- China
- Prior art keywords
- image
- foreground
- background
- deterministic
- information entropy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000003709 image segmentation Methods 0.000 title claims abstract description 24
- 238000000605 extraction Methods 0.000 title claims abstract description 18
- 230000002452 interceptive effect Effects 0.000 title claims abstract description 14
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 25
- 230000000877 morphologic effect Effects 0.000 claims abstract description 12
- 239000000203 mixture Substances 0.000 claims abstract description 9
- 230000011218 segmentation Effects 0.000 claims description 21
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000011426 transformation method Methods 0.000 claims description 5
- 230000010339 dilation Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 239000003086 colorant Substances 0.000 claims description 2
- 210000004185 liver Anatomy 0.000 abstract description 10
- 230000000694 effects Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 8
- 230000003187 abdominal effect Effects 0.000 description 6
- 210000000056 organ Anatomy 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 2
- 239000012535 impurity Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20152—Watershed segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30056—Liver; Hepatic
Abstract
The invention discloses a medical image segmentation method based on interactive foreground extraction and information entropy watershed, which comprises the following steps: carrying out standardization processing on an original image; eliminating white point noise and edges existing in the image by utilizing morphological open operation; marking the approximate position of the image foreground by using a rectangular frame, and removing a background area in the image; modeling deterministic foreground and background of the image by using a Gaussian mixture model, creating new pixel distribution, and generating a complete image; finding a threshold value of the complete image segmentation by using the image information entropy, and converting the threshold value into a binary image; and (4) carrying out image extraction on the binary image through a watershed algorithm to obtain a required image. According to the invention, the image edges are filtered by an interactive foreground extraction method, the information entropy and the watershed algorithm are combined, the image can be effectively segmented, the obtained liver CT image is complete, and the interference caused by uneven distribution of pixel values, interconnection of foreground subgraphs and different shapes of liver organs among individuals is overcome.
Description
Technical Field
The invention relates to the field of image processing, in particular to a medical image segmentation method based on interactive foreground extraction and information entropy watershed.
Background
With the development of medical imaging technology and image pattern recognition technology, image segmentation plays a leading role in the field of medical processing and analysis, and the image segmentation mainly aims to segment a part with specific significance in a medical image, so that a reliable basis is provided for clinical diagnosis and pathological research, the mechanized film reading burden of a doctor is effectively reduced, and more accurate diagnosis is made. Since the medical image has complexity, a series of problems such as individual difference and uneven distribution of pixel values need to be solved in the segmentation process. At present, a broad medical image segmentation theory and method do not exist.
At present, there are many medical image segmentation methods widely used at home and abroad, such as segmentation methods based on region growing. The region growing method is to assemble pixels with similar properties to form a region, and the specific idea is as follows: firstly, a seed pixel is selected from the target segmentation area as a starting point for growth, and then pixels with similar properties in the area around the seed pixel are merged into the area where the seed pixel is located until no adjacent point or area is available for merging. In order to improve the segmentation quality, the literature "medical image segmentation based on morphology and region growing method" proposes medical image segmentation based on morphology and region growing method, which corrects the growth result by using the basic operation of mathematical morphology, fills the small hole noise after region growing, and improves the segmentation effect. The literature, "research on OCT image segmentation algorithm based on region growing" proposes research on OCT image segmentation algorithm based on region growing, and an improved algorithm based on region growing is proposed by analyzing the properties of medical OCT images.
In general, in an abdominal CT image, pixel values are generally concentrated in a lower range, contrast between abdominal organs is lacking, and due to organ lesions and differences in sizes of organ shapes between individuals, a conventional segmentation method for finding an image edge by using a gray information value is difficult to obtain a better segmentation effect, and the processing thereof has the influence of incomplete segmentation or more noises.
Disclosure of Invention
In order to solve the technical problems, the invention provides a medical image segmentation method based on interactive foreground extraction and information entropy watershed, which is simple in algorithm, high in segmentation speed and high in segmentation precision.
The technical scheme for solving the problems is as follows: a medical image segmentation method based on interactive foreground extraction and information entropy watershed comprises the following steps:
the method comprises the following steps: carrying out standardization processing on an original image;
step two: eliminating white point noise and edges in the image after the standardization processing by utilizing morphological open operation;
step three: marking the approximate position of the image foreground by using a rectangular frame, determining the region outside the rectangular frame as a deterministic background, and removing the background region in the image, wherein the other regions are deterministic foreground;
step four: modeling deterministic foreground and background of the image by using a Gaussian mixture model, creating new pixel distribution, and generating a complete image according to the obtained pixel distribution condition;
step five: finding a threshold value of the complete image segmentation by using the image information entropy, and converting the threshold value into a binary image;
step six: and (4) carrying out image extraction on the binary image through a watershed algorithm to obtain a final required image.
In the first step, whether the original image meets the standard size is judged, if not, the original image is standardized, and then threshold processing is carried out to convert the original image into a binary image.
In the fourth step, a Gaussian mixture model is used for modeling deterministic foreground and background, the Gaussian mixture model creates new pixel distribution according to a preset position, a complete image is generated according to the obtained pixel distribution condition, each node in the image is a pixel point, all background pixels are connected with background nodes, and all foreground pixels are connected with foreground nodes; the weight of each pixel classified into a background node or a foreground node is determined by the probability of whether the pixel is a background or a foreground, the weight of the edge of two pixels is determined by the similarity between the background and the foreground at the joint between the background and the foreground, and the closer the pixel color between the background and the foreground is, the greater the weight of the edge is; dividing according to the weight relation of each edge, and dividing different pixel points into foreground nodes and background nodes; the above process is repeated until the classification converges.
In the fifth step, the complete image obtained in the fourth step of image maximum information entropy segmentation is adopted; the specific process is as follows:
5-1) in the image, the probability of a certain gray value x is set as p (x), and then the solving formula of the information entropy H is shown as (1):
5-2) in the gray image, if a certain threshold value is set as T, the gray value in the range of 0-T is regarded as background B, and the gray value in the range of T +1 to T +255 is regarded as foreground F; calculating the probability of each gray value in the background and the foreground, wherein the calculation formulas are respectively shown as (2) and (3):
wherein p isiRepresenting the probability of a grey value of i, PBRepresenting the probability, P, of each gray level in the background BFRepresenting the probability, p, of each gray level in the foreground FTRepresents the probability of the threshold T in the entire image;
5-3) calculating background information entropy HBWith foreground information entropy HFThe calculation formulas are shown in (4) and (5), respectively.
And (3) traversing the exhaustive threshold value to be 0-255, and obtaining the maximum threshold value T when the image has the maximum information entropy, wherein the threshold value is the optimal threshold value point for segmenting the gray level image.
In the sixth step, for a gray image, the gray image is regarded as a terrain on a three-dimensional geographic map, the region with lower gray value is regarded as a valley, the region with higher gray value is regarded as a peak, and the watershed algorithm implementation process is to divide the image into two different sets: watershed lines and ponding basins; the watershed algorithm comprises the following specific processes:
6-1) firstly, carrying out morphological open operation on the image to eliminate noise in the image;
6-2) because the sub-images in the image are connected together, extracting the deterministic foreground by adopting a distance transformation method;
6-3) processing the image threshold obtained by distance transformation to obtain the deterministic foreground of the actual image;
6-4) after obtaining the deterministic foreground image, obtaining a deterministic background, and performing morphological dilation processing on the image after opening operation, wherein the background of the obtained dilated image is the deterministic background of the original image, and the dilated image minus the deterministic foreground image is an unknown area;
6-5) dividing the whole image space according to three major classes of a deterministic background, a deterministic foreground and an unknown region, wherein the deterministic background is represented by '1', the deterministic foreground is represented by positive integers from '2', the unknown region is represented by '0', and each class is labeled by different colors;
6-6) segmenting the image by utilizing a watershed algorithm, and dividing regions of different classes.
In the step 6-2), the distance from a non-zero-value pixel point in the image to the nearest zero-value pixel point is calculated by adopting a distance transformation method, that is, the distance between all pixel points in the binary image and the nearest pixel point with the value of 0 is calculated.
The invention has the beneficial effects that: according to the invention, the image edges are filtered by an interactive foreground extraction method, the image can be effectively segmented by combining the information entropy and the watershed algorithm, the obtained liver CT image is complete and has no more impurity interference, and the interference caused by uneven distribution of pixel values, interconnection of foreground subgraphs and different shapes of liver organs among individuals is overcome.
Drawings
FIG. 1 is a flow chart of the present invention.
FIG. 2 is a diagram illustrating threshold processing and morphological operation transitions in accordance with an embodiment of the present invention.
Fig. 3 is a process diagram of filtering out deterministic backgrounds in an embodiment of the invention.
Fig. 4 is a process diagram of extracting a complete abdominal organ image according to an embodiment of the present invention.
FIG. 5 is a diagram illustrating the effect of entropy processing of image information according to an embodiment of the present invention.
Fig. 6 is a diagram illustrating the extracting effect of the deterministic foreground in the embodiment of the present invention.
FIG. 7 is a diagram of a deterministic background and unknown regions in an embodiment of the invention.
FIG. 8 is a diagram illustrating an effect of dividing an image space according to an embodiment of the present invention.
FIG. 9 is a diagram illustrating the effect of segmentation using a watershed algorithm according to an embodiment of the present invention.
Fig. 10-13 are graphs showing the results of the segmentation operation performed on 4 different images according to the embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings and examples.
As shown in fig. 1, a method for segmenting a medical image based on interactive foreground extraction and information entropy watershed includes the following steps:
the method comprises the following steps: and carrying out standardization processing on the original image.
Firstly, judging whether the original image meets the standard size, if not, standardizing the original image, and then converting the original image into a binary image through threshold processing.
Step two: and eliminating white point noise and fine edges in the obtained binary image by utilizing morphological open operation, thereby realizing the preliminary elimination of the edges. The conversion result is shown in fig. 2.
Step three: after the abdominal CT image is subjected to threshold processing and morphological operation, the approximate position of the image foreground can be marked by using a rectangular frame, the region outside the rectangular frame is determined as a deterministic background, and other regions are determined as deterministic foregrounds, so that the background region in the image is preliminarily removed, and a good template is provided for a subsequent segmentation task. The operation is shown in fig. 3.
Step four: and modeling the deterministic foreground and background of the image by using a Gaussian mixture model, creating new pixel distribution, and generating a complete image according to the obtained pixel distribution condition.
Since the image to be processed has been size-normalized at step one, the differences between individuals can be largely eliminated. The algorithm can preset the approximate locations of the deterministic foreground and the deterministic background of the image. The deterministic foreground and background are modeled by a Gaussian mixture model, which creates a new pixel distribution according to a preset position and classifies unclassified pixels (which may be background or foreground) in sequence according to their relationship to preset classification pixels (foreground and background). Generating a complete image according to the obtained pixel distribution condition, wherein each node in the image is a pixel point, all background pixels are connected with the background nodes, and all foreground pixels are connected with the foreground nodes; the weight of each pixel classified into a background node or a foreground node is determined by the probability of whether the pixel is a background or a foreground, the weight of the edge of two pixels is determined by the similarity between the background and the foreground at the joint between the background and the foreground, and the closer the pixel color between the background and the foreground is, the greater the weight of the edge is; dividing according to the weight relation of each edge, and dividing different pixel points into foreground nodes and background nodes; the above process is repeated until the classification converges. An image of the entire abdominal organ was extracted as shown in fig. 4.
Step five: and finding a threshold value of the complete image segmentation by using the image information entropy, and converting the threshold value into a binary image.
The threshold segmentation divides the gray level image into one or more areas based on the selection of the gray level threshold, so as to realize the segmentation of the image. The key of the threshold segmentation method is a threshold selection algorithm, and the commonly used algorithm comprises the maximum between-class variance (OTSU method) and the information entropy. In order to improve the segmentation precision of the liver organ, a complete image obtained in the fourth image maximum information entropy segmentation step is adopted; the specific process is as follows:
5-1) in the image, the probability of a certain gray value x is set as p (x), and then the solving formula of the information entropy H is shown as (1):
5-2) in the gray image, if a certain threshold value is set as T, the gray value in the range of 0-T is regarded as background B, and the gray value in the range of T +1 to T +255 is regarded as foreground F; calculating the probability of each gray value in the background and the foreground, wherein the calculation formulas are respectively shown as (2) and (3):
wherein p isiRepresenting the probability of a grey value of i, PBRepresenting the probability, P, of each gray level in the background BFRepresenting the probability, p, of each gray level in the foreground FTRepresents the probability of the threshold T in the entire image;
5-3) calculating background information entropy HBWith foreground information entropy HFThe calculation formulas are shown in (4) and (5), respectively.
And (3) traversing the exhaustive threshold value to be 0-255, and obtaining the maximum threshold value T when the image has the maximum information entropy, wherein the threshold value is the optimal threshold value point for segmenting the gray level image. The effect is shown in fig. 5. When the threshold segmentation is realized by using the information entropy, the optimal threshold can be found by adopting the maximum between-class variance (OTSU method).
Step six: and (4) carrying out image extraction on the binary image through a watershed algorithm to obtain a final required image.
Regarding a gray image as a terrain on a three-dimensional geographic map, regarding an area with lower gray value as a valley, regarding an area with higher gray value as a peak, and dividing the image into two different sets by a watershed algorithm implementation process: watershed lines and ponding basins; the watershed algorithm comprises the following specific processes:
6-1) firstly, carrying out morphological open operation on the image, and eliminating noise in the image so as to avoid the influence of the noise on image segmentation.
6-2) since the intra-image subgraphs are usually connected together, the deterministic foreground needs to be extracted by means of distance transformation. In general, the distance from a non-zero-value pixel point in an image to the nearest zero-value pixel point is calculated by means of a distance transformation method, that is, the distance between all pixel points in a binary image and the nearest pixel point with the value of 0 is calculated. The process not only can realize the preliminary acquisition of the image deterministic foreground, but also can calculate the center of a foreground object, refine the outline and the like.
6-3) then threshold the distance transformed image to obtain the deterministic foreground of the actual image, as shown in FIG. 6.
6-4) after obtaining the deterministic foreground image, obtaining a deterministic background, performing morphological dilation processing on the image after opening operation, wherein the background of the obtained dilated image is the deterministic background of the original image, and the dilated image minus the deterministic foreground image is an unknown area; the effect is shown in fig. 7.
6-5) dividing the whole image space according to three major classes of a deterministic background, a deterministic foreground and an unknown area, wherein the deterministic background is represented by '1', the deterministic foreground is represented by positive integers from '2', the unknown area is represented by '0', and each class is marked by gray, black and white, and the effect is shown in figure 8.
6-6) segmenting the image by utilizing a watershed algorithm, and dividing regions of different classes. Since the liver positions are approximately the same among individuals, the liver image can be screened according to the approximate positions. The effect is shown in fig. 9.
The liver image segmentation operation is performed on 4 different images by using the algorithm of the present invention, and the results are shown in fig. 10, fig. 11, fig. 12 and fig. 13, respectively. As can be seen from the image segmentation result, the image edge is filtered by the interactive foreground extraction method, and the image can be effectively segmented by combining the information entropy and the watershed algorithm. The obtained liver CT image is complete and has no more impurity interference, and the interference caused by uneven distribution of pixel values, interconnection of foreground subgraphs and different shapes of liver organs among individuals is overcome. The method has good image segmentation effect when a large number of medical abdominal CT images are segmented.
Claims (6)
1. A medical image segmentation method based on interactive foreground extraction and information entropy watershed is characterized by comprising the following steps:
the method comprises the following steps: carrying out standardization processing on an original image;
step two: eliminating white point noise and edges in the image after the standardization processing by utilizing morphological open operation;
step three: marking the approximate position of the image foreground by using a rectangular frame, determining the region outside the rectangular frame as a deterministic background, and removing the background region in the image, wherein the other regions are deterministic foreground;
step four: modeling deterministic foreground and background of the image by using a Gaussian mixture model, creating new pixel distribution, and generating a complete image according to the obtained pixel distribution condition;
step five: finding a threshold value of the complete image segmentation by using the image information entropy, and converting the threshold value into a binary image;
step six: and (4) carrying out image extraction on the binary image through a watershed algorithm to obtain a final required image.
2. The method as claimed in claim 1, wherein in the step one, it is first determined whether the original image meets a standard size, and if not, the original image is normalized, and then the original image is converted into a binary image by threshold processing.
3. The method for segmenting the medical image based on the interactive foreground extraction and the information entropy watershed as claimed in claim 1, wherein in the fourth step, a gaussian mixture model is used for modeling the deterministic foreground and the background, the gaussian mixture model creates a new pixel distribution according to a preset position, and a complete image is generated according to the obtained pixel distribution condition, each node in the image is a pixel point, all background pixels are connected with the background nodes, and all foreground pixels are connected with the foreground nodes; the weight of each pixel classified into a background node or a foreground node is determined by the probability of whether the pixel is a background or a foreground, the weight of the edge of two pixels is determined by the similarity between the background and the foreground at the joint between the background and the foreground, and the closer the pixel color between the background and the foreground is, the greater the weight of the edge is; dividing according to the weight relation of each edge, and dividing different pixel points into foreground nodes and background nodes; the above process is repeated until the classification converges.
4. The interactive foreground extraction and information entropy watershed-based medical image segmentation method of claim 3, wherein in the fifth step, the complete image obtained in the fourth step of image maximum information entropy segmentation is adopted; the specific process is as follows:
5-1) in the image, the probability of a certain gray value x is set as p (x), and then the solving formula of the information entropy H is shown as (1):
5-2) in the gray image, if a certain threshold value is set as T, the gray value in the range of 0-T is regarded as background B, and the gray value in the range of T +1 to T +255 is regarded as foreground F; calculating the probability of each gray value in the background and the foreground, wherein the calculation formulas are respectively shown as (2) and (3):
wherein p isiRepresenting the probability of a grey value of i, PBRepresenting the probability, P, of each gray level in the background BFRepresenting the probability, p, of each gray level in the foreground FTRepresents the probability of the threshold T in the entire image;
5-3) calculating background information entropy HBWith foreground information entropy HFThe calculation formulas are shown in (4) and (5), respectively.
And (3) traversing the exhaustive threshold value to be 0-255, and obtaining the maximum threshold value T when the image has the maximum information entropy, wherein the threshold value is the optimal threshold value point for segmenting the gray level image.
5. The method as claimed in claim 4, wherein in the sixth step, for a gray image, it is regarded as a terrain on a three-dimensional geographic map, the areas with lower gray values are regarded as valleys, the areas with higher gray values are regarded as peaks, and the watershed algorithm is implemented by dividing the image into two different sets: watershed lines and ponding basins; the watershed algorithm comprises the following specific processes:
6-1) firstly, carrying out morphological open operation on the image to eliminate noise in the image;
6-2) because the sub-images in the image are connected together, extracting the deterministic foreground by adopting a distance transformation method;
6-3) processing the image threshold obtained by distance transformation to obtain the deterministic foreground of the actual image;
6-4) after obtaining the deterministic foreground image, obtaining a deterministic background, and performing morphological dilation processing on the image after opening operation, wherein the background of the obtained dilated image is the deterministic background of the original image, and the dilated image minus the deterministic foreground image is an unknown area;
6-5) dividing the whole image space according to three major classes of a deterministic background, a deterministic foreground and an unknown region, wherein the deterministic background is represented by '1', the deterministic foreground is represented by positive integers from '2', the unknown region is represented by '0', and each class is labeled by different colors;
6-6) segmenting the image by utilizing a watershed algorithm, and dividing regions of different classes.
6. The method for segmenting the medical image based on the interactive foreground extraction and the information entropy watershed as claimed in claim 5, wherein in the step 6-2), the distance from the non-zero-value pixel point in the image to the nearest zero-value pixel point is calculated by adopting a distance transformation method, namely, the distance from all the pixel points in the binary image to the nearest pixel point with the value of 0 is calculated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011226634.0A CN112330561B (en) | 2020-11-05 | 2020-11-05 | Medical image segmentation method based on interactive foreground extraction and information entropy watershed |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011226634.0A CN112330561B (en) | 2020-11-05 | 2020-11-05 | Medical image segmentation method based on interactive foreground extraction and information entropy watershed |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112330561A true CN112330561A (en) | 2021-02-05 |
| CN112330561B CN112330561B (en) | 2023-06-13 |
Family
ID=74315458
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011226634.0A Active CN112330561B (en) | 2020-11-05 | 2020-11-05 | Medical image segmentation method based on interactive foreground extraction and information entropy watershed |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112330561B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113178002A (en) * | 2021-04-26 | 2021-07-27 | 天津知北科技有限公司 | 3D picture display method, device and equipment based on mobile terminal |
| CN113506246A (en) * | 2021-06-15 | 2021-10-15 | 西安建筑科技大学 | Concrete 3D printing component fine detection method based on machine vision |
| CN116030633A (en) * | 2023-02-21 | 2023-04-28 | 天津汉云工业互联网有限公司 | Vehicle tunnel early warning method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103632361A (en) * | 2012-08-20 | 2014-03-12 | 阿里巴巴集团控股有限公司 | An image segmentation method and a system |
| US20150049954A1 (en) * | 2012-03-19 | 2015-02-19 | Satoshi Hikida | Image processing apparatus, image processing method, and computer-readable recording medium |
| CN104715474A (en) * | 2015-01-20 | 2015-06-17 | 电子科技大学 | High-resolution synthetic aperture radar image linear building detecting method based on marked watershed algorithm |
| CN104881669A (en) * | 2015-05-13 | 2015-09-02 | 中国科学院计算技术研究所 | Method and system for extracting local area detector based on color contrast |
| CN110335278A (en) * | 2019-05-16 | 2019-10-15 | 陕西师范大学 | A method of the colloid direction character of quantization tumor microenvironment |
-
2020
- 2020-11-05 CN CN202011226634.0A patent/CN112330561B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150049954A1 (en) * | 2012-03-19 | 2015-02-19 | Satoshi Hikida | Image processing apparatus, image processing method, and computer-readable recording medium |
| CN103632361A (en) * | 2012-08-20 | 2014-03-12 | 阿里巴巴集团控股有限公司 | An image segmentation method and a system |
| CN104715474A (en) * | 2015-01-20 | 2015-06-17 | 电子科技大学 | High-resolution synthetic aperture radar image linear building detecting method based on marked watershed algorithm |
| CN104881669A (en) * | 2015-05-13 | 2015-09-02 | 中国科学院计算技术研究所 | Method and system for extracting local area detector based on color contrast |
| CN110335278A (en) * | 2019-05-16 | 2019-10-15 | 陕西师范大学 | A method of the colloid direction character of quantization tumor microenvironment |
Non-Patent Citations (2)
| Title |
|---|
| 张海涛等: "阈值标记的分水岭彩色图像分割", 《中国图象图形学报》 * |
| 曾宪华等: "多特征融合的医学影像颜色感知算法", 《计算机辅助设计与图形学学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113178002A (en) * | 2021-04-26 | 2021-07-27 | 天津知北科技有限公司 | 3D picture display method, device and equipment based on mobile terminal |
| CN113506246A (en) * | 2021-06-15 | 2021-10-15 | 西安建筑科技大学 | Concrete 3D printing component fine detection method based on machine vision |
| CN113506246B (en) * | 2021-06-15 | 2022-11-25 | 西安建筑科技大学 | Concrete 3D printing component fine detection method based on machine vision |
| CN116030633A (en) * | 2023-02-21 | 2023-04-28 | 天津汉云工业互联网有限公司 | Vehicle tunnel early warning method and device |
| CN116030633B (en) * | 2023-02-21 | 2023-06-02 | 天津汉云工业互联网有限公司 | Vehicle tunnel early warning method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112330561B (en) | 2023-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112330561B (en) | Medical image segmentation method based on interactive foreground extraction and information entropy watershed | |
| CN111145209B (en) | Medical image segmentation method, device, equipment and storage medium | |
| WO2019062092A1 (en) | Superpixel- and multivariate color space-based body outline extraction method | |
| CN109978848B (en) | Method for detecting hard exudation in fundus image based on multi-light-source color constancy model | |
| CN111882561A (en) | Cancer cell identification and diagnosis system | |
| CN106780728B (en) | Medical image-based single organ splitting method and system | |
| CN108399644A (en) | A kind of wall images recognition methods and its device | |
| CN104766316A (en) | Novel lip segmentation algorithm for traditional Chinese medical inspection diagnosis | |
| CN108986109B (en) | Automatic segmentation method for serialized visible human body slice images | |
| Lai et al. | Effective segmentation for dental X-ray images using texture-based fuzzy inference system | |
| CN112102332A (en) | Cancer WSI segmentation method based on local classification neural network | |
| JP4285640B2 (en) | Object identification method, apparatus and program | |
| CN116824168A (en) | Ear CT feature extraction method based on image processing | |
| JP2009123234A (en) | Object identification method, apparatus and program | |
| CN108921179B (en) | Automatic extraction and quantification method for colors of lesion areas of infantile hemangioma | |
| CN108629780B (en) | Tongue image segmentation method based on color decomposition and threshold technology | |
| CN111681185B (en) | Finite element modeling method based on X-ray scanning image of asphalt mixture | |
| CN109191481B (en) | Rapid threshold segmentation method for shoe print image under poor exposure condition | |
| CN109934215B (en) | Identification card identification method | |
| Sreejini et al. | M-FISH karyotyping-a new approach based on watershed transform | |
| CN113744241A (en) | Cell image segmentation method based on improved SLIC algorithm | |
| Fedorov et al. | Development of technique for face detection in image based on binarization, scaling and segmentation methods | |
| CN112712541A (en) | Image shape descriptor generation method | |
| Devi et al. | Detection of tumours from MRI scans using Segmentation techniques | |
| CN112102350A (en) | Secondary image segmentation method based on Otsu and Tsallis entropy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |